New Delhi, Sep 16 : While climate change is making the pattern of rainfall more erratic and difficult to predict, the Indian government has started providing AI-driven forecasts directly to farmers on smartphones to help them manage their agricultural operations.

'This year's unusual monsoon, with its early arrival and a rare midseason hiatus, illustrates just how fragile agricultural rhythms have become. But this year was also different for another reason: millions of Indian farmers had artificial intelligence on their side,' according to an article by Stacey Glaser in One World Outlook.

Earlier, accurate weather forecasting could be done only by governments and well-funded institutions, backed by supercomputers that cost millions of dollars.

'But AI has begun to dismantle that exclusivity. With open-source models like Google's NeuralGCM and the European Center for Medium-Range Weather Forecasts' AI systems, highly granular predictions are now possible on devices as modest as a farmer's smartphone,' according to an article by Stacey Glaser in One World Outlook.

The article refers to this wider availability of forecasting as not just a technological innovation but a political and social breakthrough. In India, the government sent AI-driven forecasts to 38 million small farmers this monsoon season, it states.

'Rather than offering broad weather summaries, forecasts were tailored to the needs of individual farmers: whether to plant early, buy more seed, or prepare for drought-like pauses,' the article states.

Weather forecasting, once dominated by elite institutions, is now being reframed as a public good. Researchers at the University of Chicago, working with the Indian government, helped bridge the gap between machine learning outputs and actionable advice.

The article cites Amir Jina, an assistant professor involved in the project, as saying, 'What dots hadn't been connected before was this tailoring of forecast to purpose.'

This tailoring is important as a national-level forecast issues an overall warning of heavy rains. However, farmers need specific information on whether they should delay planting rice seedlings or whether a sugarcane field should be irrigated. This requires locally relevant information, which is being made possible through AI forecasts.

The article also points out that the scale of India's forecasting also opens an opportunity for emulating the same method in other countries.

'If India can deploy AI-driven forecasts at scale for some of the world's poorest farmers, then other developing nations may follow suit. In the face of climate change, information is not just power -- it is survival,' the article added.

/IANS